Web Analytics

Core Principles to Get Started with Conversion Rate Optimization

amol23813383 views

I don’t care whether you use Google Website Optimizer or Omniture. The principles below are universal and it doesn’t matter what software you use for testing. Only 2 conditions need to be met in order for this article to be relevant to you:

  • You need to be willing to test different variations of your original landing page
  • You need to be somewhat familiar with this whole concept of conversion rate optimization, know how to set up a basic test, how to make sure your results are valid (if you’re using Google Website Optimizer they do this for you)

If you regularly read Web Traffic ROI, then I’m sure you already know these 2 things. I’m also pretty sure you’ve read some of the author’s great free eBook (you can do it by entering your email on the right side of this screen). Anyway, let’s get started:

Start With Making Major Changes to Your Landing Page

Get started by testing DRASTICALLY different versions of your landing page. By drastically different I don’t mean changing the color of the ‘buy now’ button! Or changing the buy now button itself.

Start by changing the whole layout and the whole structure of the page. Want an example?

Compare this:

to this:

(by the way, these examples are from WhichTestWon, a nice website I found while doing a research for this article)

Can you guess which page produced more conversions? It was the first variation, resulting in 60% better conversion rate although the visitors had to make an extra click. Our intuitions are not perfect in this area and that’s why we need to test and test often.

Do you think those guys would get a same conversion rate increase if they just changed the sign up button? Or maybe the headline text? I doubt so. What they did instead is re-organizing the whole page and making some drastic changes.

When you start with doing split tests, don’t preoccupy yourself with doing small changes that will probably produce small results. Instead, do some bigger changes. Use your imagination. Change the title, image, call to action and the credibility indicators in the same time (change their appearance and their location on the screen). See what happens. With big changes you either get big losses or big wins. We sure hope for the second option here.

After You Found a Winning Landing Page, Test Small Changes

Finally found the perfect landing page that converts 1034350% better than the original? 🙂 Now it’s time to do some smaller tests. Try changing the headline only or maybe the ‘buy’ now button.

Wait, why should we do small changes now? Why not continue with major changes only until we get a MAJOR win?

Because it’s not all about what converts but also WHY it converts. With small changes you can find what exactly is the factor that drives a significant conversion rate increase. Maybe it was that your visitors were anxious to purchase and needed some credibility signs like ‘verified by _____’ and so on.

Now, imagine for a moment this is true. You can now use this knowledge and add credibility signs to ALL of your pages (the shopping cart, the ‘product description’ pages) and have a big probability to see an increase in conversions on all of those pages! That’s the power of understanding why something happens, because you can then take that thing and apply it to other areas as well.

Do These Principles Apply Only for Landing Page Optimization?

No! These principles are useful for all types of optimization. Let me give you an example.

This article on people search for free is currently the most popular article on my site, FinderMind.com. As you can see, I use AdSense to monetize the page. Several weeks ago I tried to do various tests to see what produced the biggest CTR. I’ve tried putting ads below the post, on the left sidebar and so on. You noticed something? I tried to make a major change with first changing the POSITION of the ad. I didn’t first try to stick to a particular position and change only the color.

Back to the test. I found that putting ads below the post title has a way bigger CTR than the rest of the positions. Now I went and did some small tests. I tested different sizes and colors till I got to where I am now.

These principles don’t just apply to landing page optimization but to ANY type of online optimization.


Related posts

How to Measure Cross Device Ad conversions on Google

7 Tips: How Online Marketers Can Benefit From Google Analytics

The Absolute Beginner’s Guide to A/B Testing

13 comments

eCommerce Development December 20, 2010 - 12:55 am
Conversion rate is an essential part for ecommerce website owners..this can be measured by web analytics of google and others. It happens that you got huge traffic to your website but what if it doesnot convert into to sales or whatever your goal is.
Tej Kohli December 20, 2010 - 9:11 am
Well here I am using both Multivariate testing and A/B testing both with my Adwords account and my ads is fully optimized but still unable to increase ROI , Will keep in mind all tricks you mentioned here
Darren December 21, 2010 - 5:40 am
Just try doing very different versions of the same page first.
Property Marbella December 20, 2010 - 9:17 am
Hi Darren, The big different are that people don’t click on “Sign up New” if they don’t know what they must sign in with. You gave all that information direct on the first page and that is more security for many people.
Alex@Jocuri December 20, 2010 - 12:00 pm
I think i said it on some other posts on this blog, that I am a strong adept of split test testing. You can win so much just by trying different version. You would be amazed on how much the conversion rates can change just by changing the position of your opt-in form or changing the background colors, but some people are just to lazy or don't think that something so small can help them.
Darren December 21, 2010 - 5:41 am
@Alex Agreed.
Aluminum Cases December 20, 2010 - 5:20 pm
Yes you can never underestimate the power of conversation rate optimization. While you may think your landing page looks great, your visitors may have major problems with it. You should always be testing things to try to squeeze more conversions out of your traffic. You may want to start by analyzing what your more successful competitors are doing differently. If those competitors are established, they may have already done conversion rate testing which you can use to your advantage.
Bubble Cove December 21, 2010 - 5:40 pm
I agree with Alex, split testing can help greatly... however it can be a bit of a pain to do! Coversion is key... especially on shop fronts. Say you pay for advertising?
Jasmine December 22, 2010 - 11:14 am
Oh my my, these are great tips you are sharing here. Looks like I need to make changes to my pages also... but I am a little lost. Not too sure how I should go about. Perhaps you can take a look and help me a bit? A tip or two will do. :) Thanks, Darren.
Darren December 22, 2010 - 11:24 am
Just listen to some of the next marketingexperiments.com webinars, every webinar they have a live optimization part where you can submit your website and get advice from their experts.
Software Development Company December 23, 2010 - 8:25 am
Split testing is best strategy to know about which is effective landing page.
Compare Car Rental January 4, 2011 - 12:08 am
Great fantastic tips. We will doing some multi variant testing soon, which is going to hopefully give us some great insights as to what people actually do when they visit our webiste. One additional tip I have test and proves works, is that on search boxes, always use words like "Go" rather than search, as they suggest to users that no-more work from them is needed. Come from the book, don't make me think! A fantastic CRO book!
Ron Teners January 25, 2011 - 8:34 pm
A good example of this optimization can be seen at AdsMain Affiliate Network. It already showing good results.
Add Comment